André Kelpe
Principal Software Engineer at Salesforce
Berlin, Germany
Summary
👤
SeniorAndré Kelpe is a Principal Software Engineer based in Berlin with 17 years of experience building distributed data systems and cloud-native back ends. He specializes in Hadoop and the JVM ecosystem—Java, Gradle, Maven—plus Big Data frameworks like Cascading, Scalding and Cascalog, and operational tooling on AWS and Elasticsearch. At Salesforce and as founder of his own consultancy he blends hands-on engineering with architecture and delivery, driving reliable data processing at scale. A long-time open-source contributor, he has made substantive fixes and compatibility work to foundational projects such as Cascading, Scalding and Cascalog, improving Hadoop integration and test robustness. Known for pragmatic TDD, CI-driven workflows and GIS/navigation expertise, he often surfaces subtle edge-case fixes that make large-scale pipelines more fault tolerant.
17 years of coding experience
English, German, Dutch
Stackoverflow
13,119reputation
11.9mreached
73answers
5questions
python
top-1%
java
top-5%
oop
top-5%
indentation
top-5%
Github Skills (28)
hadoop-mapreduce10
cascading10
data-pipelines10
python10
clojure-cli10
mapreduce10
hadoop10
java10
scala10
clojure10
javas10
project-configuration10
data-processing10
data-pipeline10
apache-hadoop9
Programming languages (22)
Github contributions (5)
cwensel/cascading
Oct 2013 - Jan 2016
Cascading is a feature rich API for defining and executing complex and fault tolerant data processing flows locally or on a cluster.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:77 commits, 2 PRs, 12 comments in 2 years 3 months
Contributions summary:André primarily contributed to the core logic of the Cascading library by addressing various issues related to file name handling, test settings, and testing order dependencies. They focused on improving the reliability and compatibility of the library, particularly in relation to Hadoop environments. Further contributions included enhancements for error reporting in Hadoop standalone mode and fixing problems related to the handling of null values and other edge cases in assembly operations. The contributions show a focus on ensuring the library functions correctly and handles different data scenarios, including those with Hadoop integration.
fault-toleranthadoopjavamapreducetez
nathanmarz/cascalog
Jun 2013 - Oct 2014
Data processing on Hadoop without the hassle.
Role in this project:Back-end Developer
Back-end DeveloperContributions:15 commits in 1 year 3 months
Contributions summary:André primarily focused on improving the Cascalog project, which appears to be a data processing framework built on Clojure and Hadoop. They enhanced the project's compatibility and maintainability by modifying the build files and configurations. A key area of contribution involved adapting the project to work with different versions of Hadoop and Cascading, suggesting a focus on ensuring the framework's robustness. The user also updated dependencies and configuration settings to improve test speeds and compatibility.
hadoop
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.